Homework 2

Loading Libraries

library(tidyverse)

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──

## ✓ ggplot2 3.3.5     ✓ purrr   0.3.4
## ✓ tibble  3.1.4     ✓ dplyr   1.0.7
## ✓ tidyr   1.1.3     ✓ stringr 1.4.0
## ✓ readr   2.0.1     ✓ forcats 0.5.1

## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()

Formatting plots

library(tidyverse)

knitr::opts_chunk$set(
  fig.width = 22,
  fig.asp = .8,
  out.width = "90%")

theme_set(theme_minimal() + theme(legend.position = "bottom"))

options(
  ggplot2.continuous.colour = "viridis",
  ggplot2.continuous.colour = "viridis")

sclae_colour_discrete = scale_color_viridis_d
scale_fill_discrete = scale_fill_viridis_d

Problem 1

library(p8105.datasets)
data("instacart")

Instacart Dataset Descriptions

• The Instacart dataset includes 15 variables: order_id, product_id, add_to_cart_order, reordered, user_id, eval_set, order_number, order_dow, order_hour_of_day, days_since_prior_order, product_name, aisle_id, department_id, aisle, department.
• The dataset has 1384617 rows and 15 columns.
• The range of days since the prior order is (0, 30).
• The median number of items added to cart order is (7).
• The total number of times an item was reordered is 828824.

Aisles: Count and Most Ordered

instacart %>% 
  summarize(
    distinct_aisles = n_distinct(aisle))  

## # A tibble: 1 × 1
##   distinct_aisles
##             <int>
## 1             134

• There are 134 different aisles.

Creating a plot

instacart %>% 
  count(aisle) %>% 
  filter(n > 10000) %>%
  mutate(
    aisle = factor(aisle),
    aisle = fct_reorder(aisle, n)) %>% 
  ggplot(aes(x = n, y = aisle)) +
  geom_point() +
  labs(
    title = "Number of Items Orders by Aisle",
    x = "Number of Items Sold",
    y = "Item"
  )

* The aisles where the most items are ordered from are 83 for fresh vegetables, 24 for fresh fruits, and 123 for packages vegetables fruits.

Instacart: Cleaning and filter

popular_items = instacart %>% 
  filter(aisle %in% c("baking ingredients", "dog food care", "packaged vegetables fruits")) %>% 
  group_by(aisle) %>% 
  count(product_name) %>% 
  mutate(rank = min_rank(desc(n))) %>% 
  filter(rank < 4)

Instacart: Table

knitr::kable(head(popular_items[, 1:4]), "simple")

aisle	product_name	n	rank
baking ingredients	Cane Sugar	336	3
baking ingredients	Light Brown Sugar	499	1
baking ingredients	Pure Baking Soda	387	2
dog food care	Organix Chicken & Brown Rice Recipe	28	2
dog food care	Small Dog Biscuits	26	3
dog food care	Snack Sticks Chicken & Rice Recipe Dog Treats	30	1

Icecream: Cleaning and filtering

applies_icecream = instacart %>% 
  filter(product_name %in% c("Pink Lady Apples", "Coffee Ice Cream")) %>% 
  group_by(product_name, order_dow) %>% 
  summarize(mean_hour = mean(order_hour_of_day)) %>% 
  pivot_wider(
    names_from = order_dow,
    values_from = mean_hour
  )

## `summarise()` has grouped output by 'product_name'. You can override using the `.groups` argument.

Icecream: Table

knitr::kable(head(applies_icecream[, 1:8]), "simple")

product_name	0	1	2	3	4	5	6
Coffee Ice Cream	13.77419	14.31579	15.38095	15.31818	15.21739	12.26316	13.83333
Pink Lady Apples	13.44118	11.36000	11.70213	14.25000	11.55172	12.78431	11.93750

Problem 2

library(p8105.datasets)
data("brfss_smart2010")

Cleaning Data

brfss_smart2010 = brfss_smart2010 %>% 
  janitor::clean_names() %>% 
  filter(topic == "Overall Health") %>% 
  drop_na(response) %>%
  mutate(response = str_to_lower(response),
        response = factor(response))

2002 states with 7 or more locations

states_2002 = brfss_smart2010 %>% 
  filter(year == 2002) %>%
  count(locationabbr) %>% 
  filter(n >= 7)

2010 states with 7 or more locations

states_2010 = brfss_smart2010 %>% 
  filter(year == 2010) %>%
  count(locationabbr) %>% 
  filter(n >= 7)

String plot of excellent responses

excellent = brfss_smart2010 %>% 
  filter(response == "excellent") %>% 
  group_by(locationdesc) %>% 
  mutate(average_data_value = mean(data_value)) %>% 
  select(year, average_data_value, locationdesc, locationabbr) %>% 
  ggplot(aes(x = year, y = average_data_value, color = locationabbr)) +
  geom_point() + geom_line(aes(group = locationabbr))

excellent

## Warning: Removed 30 rows containing missing values (geom_point).

## Warning: Removed 1 row(s) containing missing values (geom_path).

Distribution Plot

plot_data = brfss_smart2010 %>% 
  filter(locationabbr == "NY") %>% 
  filter(year == c(2006,2010)) %>% 
  ggplot(aes(x = response, y = data_value, color = response)) + 
  geom_boxplot() + 
  facet_grid(. ~ year)

## Warning in year == c(2006, 2010): longer object length is not a multiple of
## shorter object length

plot_data

Problem 3

accel = read_csv(file = "data/accel_data.csv")

## Rows: 35 Columns: 1443

## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr    (1): day
## dbl (1442): week, day_id, activity.1, activity.2, activity.3, activity.4, ac...

## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Accelerometer Dataset Cleaning

clean_accel = accel %>% 
  janitor::clean_names() %>% 
  mutate(week_day = ifelse(day %in% c("Saturday", "Sunday"), "weekday", "weekend")) %>% 
  select(week_day, everything()) %>% 
  pivot_longer(
    activity_1:activity_1440,
    names_to = "activity_day",
    values_to = "activity_counts")

Accelerometer Dataset Description

• The Accelerometer dataset includes variables week_day, week, day_id, day, and activity.
• The dataset has 35 rows and 1443 columns.
• The range of activity on day 1 is (, -).
• The total amount of activity on day 1 is 0.

Aggregating Accelerometer Dataset

clean_accel = clean_accel %>%
  group_by(day_id) %>% 
  summarize(sum_activity = sum(activity_counts)) 

knitr::kable(head(clean_accel))

day_id	sum_activity
1	480542.62
2	78828.07
3	376254.00
4	631105.00
5	355923.64
6	307094.24